Wednesday, September 24, 2014
This is the tenth in a series of posts that are part of a virtual symposium on the new book by Omri Ben-Shahar and Carl E. Schneider, More Than You Wanted to Know: The Failure of Mandated Disclosure. Biographies for the second week's contributors can be found here. The authors' introduction to the symposium can be found here.
Jeff Sovern is a Professor of Law at St. John’s University in New York City where he teaches Consumer Protection, among other subjects.
Ben-Shahar’s and Schneider’s book convincingly demonstrates that disclosures are rarely effective consumer protections. But there is one type of consumer disclosure where, in my view, at least, the jury is still out, though I think Ben-Shahar and Schneider feel differently. I refer to the single-letter restaurant grade disclosures that some states and municipalities require be posted on restaurant doors. I want to discuss those disclosures in this post, in the hope that they can also be extended to other situations.
Localities have long inspected restaurants for health code violations. In the pre-internet era, the resulting reports were often available for public inspection in government offices, but no doubt few consumers troubled to travel to those offices to read them. When the internet became available, some municipalities posted the reports on the web, and some consumers surely consulted them, but it seems likely that few bothered. As a result, there is no reason to think the inspections had much impact on consumer decisions about which restaurants to patronize. At some point, Los Angeles, later joined by New York City, San Diego, South Carolina, and various other governments, began distilling the grades down to a single letter grade—often an A,B, or a C--and requiring restaurants to display them at their entrances. Theoretically, enabling consumers to see a readily-understandable disclosure at the time they made a decision to enter the restaurant would make it possible for consumers to use the disclosure.
As Ben-Shahr and Schneider note at page 155, initial reports about restaurant grades were quite positive. Hospitalizations for food-borne illnesses in Los Angeles fell by 20% while the later adoption in New York was said to cut salmonella cases by 14%.Apparently, restaurateurs, not wanting to have a bad grade on their front door, improved hygiene. The resulting improvements benefited not only those who heeded the grades—by telling them which restaurants were less clean—but seemingly even those who ignored the grades, because restaurants became cleaner. It seemed like the perfect disclosure, in that it appeared all restaurant patrons were helped by it.
But a later study was less promising. Daniel Ho’s article, Fudging the Nudge: Information Disclosure and Restaurant Grading, 122 Yale L. J. 574 (2012), found, in Ben-Shahar’s and Schneider’s words that “grades have no discernable health benefits, distort the allocation of inspection resources, and mislead diners.” While Ben-Shahar and Schneider don’t say explicitly that single letter grading is a failure, the book left me, at least, with the impression that that was the authors’ view, which of course is in keeping with the rest of the disclosures the book describes. While Ben-Shahar & Schneider quite rightly point out genuine problems with single-letter grading, I believe that it is premature to conclude that it can’t work in some contexts. It would be unfortunate if governments abandoned a form of disclosure that might work merely because many other forms do not.
Ho’s claim that restaurant grades do not affect health is based on his findings that the grades did not have an impact on calls to 311 and Google searches. The assumption is that people suffering stomach issues from eating at unclean restaurants would use Google for relevant searches and that therefore if the grades resulted in fewer stomach ailments, we would see fewer such searches; because the searches didn’t decline, neither did the incidence of illnesses caused by eating at restaurants. Similarly, if people suffered gastrointestinal problems they attributed to eating at unclean restaurants, they could be expected to call 311 and report the restaurant; the absence of a drop in calls about restaurant therefore implies that there was not a drop in stomach problems.
That makes sense, up to a point. But all it gets us to, when the earlier studies are taken into account, is that the evidence for the effectiveness of single-letter disclosure grades is inconsistent. The bigger problem, however, is that while the studies may provide the best evidence we have available at present to evaluate single-letter disclosures, they are flawed as measures of such disclosures. Changes in public health during the relevant periods may be attributable to many things, of which the restaurant grades are only one. Obviously, people who become ill through eating foods prepared at home would not be affected by restaurant grades. But even if we ignore that, another problem is that consumers may find the disclosures useful even if the disclosures do not improve public health.
I teach in New York City, one of the localities using health department grades. For some years, when I take up the topic of disclosure in my consumer law course, I have asked students whether they take the health grade disclosures into account in choosing among restaurants. Usually a minority of students report that they do not. It is fair, I think, to say that those students do not benefit from the grades, unless the grades positively alter restaurateur conduct, which, as noted above, they may. But most students have claimed that they do take the grades into account. And their concern did not seem motivated solely by health concerns, though that was a factor. Another factor was disgust. Some students seem interested in avoiding unclean restaurants wholly apart from health concerns because they are repulsed by eating insect parts, etc.
The students I have polled are too small a sample to shed much light on consumer behavior generally, and as law students who have chosen to take a consumer law class, are hardly a random sample of the population. In addition, perhaps my students exaggerated the extent to which they paid attention to the grades. But I don’t think so. During one of the semesters, the cafeteria in our school received a poor grade, and students reported that they stopped eating there. I doubt they would have invented that claim. And because our building is some distance from the other eateries, the decision to avoid the cafeteria imposed some inconvenience, suggesting that the students in question valued cleanliness and the grades which signaled the lack thereof more than the convenience of eating in the cafeteria.
In short, even if the grades did not affect public health, about which there is some doubt, consumers could still find them valuable by enabling avoidance of behavior—such as eating food shared by rodents—that they see as repulsive. This does not answer the question of whether the grades are worth incurring the costs they impose, but it does indicate that the grades have utility.
To be sure, some existing restaurant grade systems have been flawed in implementation. Professor Ho observed, and Professors Ben-Shahar and Schneider echoed, that in San Diego, grade inflation has resulted in 99.9% of the restaurants receiving an A. In New York, they report, restaurant inspectors grading practices seem not to be consistent. But that does not mean that similar grading systems cannot be implemented well. Perhaps they can’t be, but we don’t have enough data to determine that.
It is easy to imagine other single-letter grading systems that might benefit consumers. For example, now that the Consumer Financial Protection Bureau has a public database for the receipt of complaints about credit card issuers, a system could assign a grade to credit card companies based on the percentage of their customers who had filed a complaint that was not satisfactorily resolved. The grade could be made available to consumers before they applied for a credit card. Would you like to know the grade a company had received before you applied for its credit card? And would such a system increase company incentives to respond to consumer complaints, or even try to forestall them, just as consumers feel pressure to pay even disputed debts to avoid soiling their credit reports?
More than most, law professors should be aware of the possibility that grades boiled down to a single symbol might be effective. Law schools have long bemoaned the impact of the US News rankings on applicants’ decisions where to go. The US News rankings are flawed, but they have an impact. To the extent that it is possible to create grades that are not flawed, they remain a tool in the consumer protection arsenal. Of course, the fact that grades work in one context doesn’t mean they will work in all contexts. Consumers might heed restaurant grades—if they do--because they wish to avoid illness or being disgusted and applicants to law schools might pay attention to the rankings because they perceive decisions as to where to study to be of great importance. Perhaps grades in contexts with less at stake would elicit less attention. And some matters surely cannot be reduced to a single grade, such as determinations about which mortgage offers the best terms. In that scenario, disclosure, whether in a single grade or at greater length, seems unlikely to be successful.
The fact is, we don’t know whether single letter grades work or not. A better test of their effectiveness would be to watch how consumers behave when selecting between two choices which received different grades. The data that we have is an imperfect proxy for that study, and it is far too soon to know whether they work. Ben-Shahar and Schneider are correct in much of what they say about disclosure’s limits. But it would be unfortunate to give up on single-letter grades given the evidence which suggests that they may help consumers.